Plugins

DSH 试金石：给「自改造」补上评测这一环——用一套金标准用例，把改动前 vs 改动后跑一遍、打分、对比，好就留、不好就撤。

Retrospective agent-session evaluation — grades the sessions that already happened: deterministic grade cards, regression diffs, /eval-history trends, and /eval --report straight from exported artifacts; no benchmark runs, no LLM judges.

DeepSeek Harness plugin for CVE impact retesting: fingerprint a target, judge it against a CVE's affected range, gather passive evidence, and report a verdict with confidence and limitations.

Test-insight plugin that turns repository facts and change risk into evidence-backed test plans and isolated, reviewable test drafts.

Input-box floating translation widget for DSH: translate drafts via session model, Zhipu GLM, or custom endpoints, with preview/replace/append, connectivity testing, and recent history.

Analyze the current branch diff and generate a Conventional Commits PR title, motivation/scheme/testing/risk description and a self-review checklist, write PR_DESCRIPTION.md, optionally open the PR via gh.

Paired experiments and promotion gates for DSH plugins.

DSH plugin: dsh-test-normalizer (npm package: @mkaliezz/dsh-test-runner)

The small deterministic operations an agent needs mid-task. JSON/YAML round-trip, JSON Schema validation, text diff, JWT decode, regex testing, SQL dialect transpiling, QR codes, timezone and unit conversion, page screenshots.

Community DSH plugin for constrained, auditable diePi research backtests

Honesty guardrails on the tool pipeline: blocks a coding agent from weakening tests, swallowing errors, stubbing type checks, or deleting tests through the shell — nine deterministic hooks, each backed by a planted-failure test proving the guard can fail.

Declarative multi-agent pipeline executor for DeepSeek Harness (dsh): runs a manifest of confined role dispatches with mechanical gates and a machine-written evidence ledger. dsh 声明式多角色流水线执行器：按 manifest 派发受限子代理、机械门禁、机器写证据账本。

WeChat Mini Program automated testing for DeepSeek Harness (dsh): selector-addressed actions, build-freshness gates, geometry-first assertions that work on text-only models, and screenshot image blocks on vision routes. 微信小程序自动化测试 dsh 插件：选择器寻址、构建新鲜度门、面向纯文

Lion-Skills — a developer-focused agent skill suite (in Chinese) as a DeepSeek Harness plugin: 11 workflow skills for code review, debugging, testing, naming, spec writing, and more.

Headless-browser verification tools so the agent opens the page it just built, reads the rendered DOM and computed styles, checks the console, and screenshots the result; ships a frontend-verify skill.

Human-in-the-loop tools (`human_task` / `human_task_ready_check`): pause the agent until the user performs a real-world action or returns an observation (GUI operation, game testing, visual verification, hardware checks), gated by session consent and an AFK presence check, returning a structured JSON result via a web task dialog.